Skip to main navigation
Skip to search
Skip to main content
Maastricht University Home
Support & FAQ
Home
Researchers
Publications
Activities
Press/Media
Prizes
Organisations
Datasets
Projects
Search by expertise, name or affiliation
Empirical Capacity Model for Self-Attention Neural Networks
Aki Härmä
,
Marcin Pietrasik
,
Anna Wilbik
Department of Advanced Computing Sciences
Advanced Computing Sciences
BISS-Brightlands Institute for Smart Society
Research output
:
Book/Report
›
Report
›
Academic
6
Downloads (Pure)
Overview
Fingerprint
Fingerprint
Dive into the research topics of 'Empirical Capacity Model for Self-Attention Neural Networks'. Together they form a unique fingerprint.
Sort by
Weight
Alphabetically
Keyphrases
Capacity Model
100%
Attention Neural Network
100%
Transformer Model
66%
Optimal number
33%
Training Data
33%
Training Algorithm
33%
Memorization
33%
Memory Capacity
33%
Number of Parameters
33%
Transformer
33%
Theoretical Capacity
33%
Design Task
33%
Synthetic Training Data
33%
Neural Transformer
33%
Algorithm Training
33%
Large Power Transformer
33%
INIS
capacity
100%
neural networks
100%
transformers
57%
algorithms
28%
data
28%
optimization
14%
design
14%
performance
14%
Computer Science
Neural Network
100%
Capacity Model
100%
Training Data
66%
training algorithm
33%
Memory Capacity
33%
Engineering
Shortfall
100%
Design Task
100%
Psychology
Neural Network
100%
Capacity Model
100%